About Clayton Kershaw’s Changeups
There was plenty about Clayton Kershaw’s Opening Day start that was predictable. He sat between 92 and 93 on that straight, riding fastball. He showed command of the pitch and didn’t walk anyone. He threw a fastball on his lone 3-1 count. The box score says he threw 27 sliders and got five whiffs — an excellent rate. His 15 curves got two whiffs and two outs on five swings. So a lot of Monday’s start was just vintage Kershaw. But that doesn’t mean he didn’t show us something a bit different.
Once he got the big lead against the Padres, out came a pitch he’s joked about working on for 10 years. Take a look at his horizontal movement against his velocity, and at least a couple dots on the plot stand out.
The raw PITCHf/x data doesn’t indicate that Kershaw threw any changeups against the Padres, but see those 85 mph pitches with arm-side movement? That’s what those are, most likely. And there’s a dot or two above them between 85 and 90 that might be changeups, too. According to Brooks Baseball, Kershaw threw only six changeups all of last year. Now it’s possible he’s thrown nearly that many on Opening Day. This requires further investigation.
There are four pitches that featured notable arm-side movement and traveled between 80 and 91 mph. Let’s order them by how impressive they were.
Fifth inning, 2 outs, 0 on, 0-1 count to right-hander Christian Bethancourt, following a 94 mph four-seamer. This one went 84 mph with 6.9 inches of fade compared to his fastball.
Sixth inning, 2 outs, 0 on, 0-1 count to right-hander Wil Myers, following a 71 mph curveball. This one went 86 mph with 4.8 inches of fade compared to his fastball.
Second inning, 0 outs, 0 on, 0-0 count to right-hander Austin Hedges. This one went 91 mph with 4.6 inches of fade compared to his fastball.
Sixth inning, 0 outs, 0 on, 0-0 count to left-hander Travis Jankowski. This one went 91 mph with 3.4 inches fade compared to his fastball.
Okay, let’s handle the latter two pitches first. One was thrown inside to a lefty (Jankowski), and though there’s some evidence that pitchers are going same-handed on changeups more often, the pitch itself isn’t in a location that normally features changeups. And it doesn’t really look like a changeup, even if it has more arm-side movement than his usual four-seam. Maybe he’s throwing a two-seamer? That pitch against Hedges also looks more like a two-seam fastball than a changeup. So probably nothing to see here.
Still, the other two pitches are characteristic of a real changeup. Which is strange, because (as noted above) Brooks Baseball said he threw six all of last year. It’s noteworthy that Kershaw threw even two of them in one start. Last year, he did that once against Atlanta in April and once before that, against the Padres (!), in his Opening Day start (!).
As for the pitch itself, it’s a reminder that we should define pitches off of the fastball rather than intrinsically. That good changeup had almost eight inches less drop than the average changeup from a lefty. It also had an inch less fade than your average lefty change. But if you compare those to his straight, riding fastball, it had nearly three inches of drop and six inches of fade — numbers that are just a little below average and almost twice average, respectively.
If Kershaw adds a changeup that looks like the one against Bethancourt, batters are screwed. He’s already proven that his breaking-ball mix can get righties out, but they’ve still managed a tiny platoon advantage against him, averaging a nutty-low .255 weighted on base average, still higher than lefties’ .239 wOBA. A changeup could neutralize the only advantage they’ve got.
By results, Kershaw had the best slider thrown by a starter last year, a top-10 curveball, and a top-15 four-seamer. Imagine adding a plus changeup.
Maybe Kershaw has made it a tradition to treat Opening Day starts against the Padres as extended Spring Training starts, or maybe the changeup is just in his head from all that spring work. From the second changeup, we can tell it’s still a work in progress. But, if you’d like to dream about dominance like we’ve never seen, replay that changeup to Bethancourt again.
In fact, let’s end on that note:
With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.
I remember seeing that one against Myers and thinking he hung that fastball really bad and was lucky to get away with it. Now I feel better about it.
He’s saving it for a few seasons from now when his fastball velocity starts ticking downward. One of years he’s going to come out of Spring Training with a junkball repertoire that would put Jamie Moyer to shame.
Just because he can.
Changeup this year. How about a knuckleball next CK? 93ish fastball plus a couple ticks from lower arm slot… high 80s slider, 70s curve, mid 80’s change, and 60’s knuckleball? Wonderful dreams. Just adding a change is great. Happy to be a Dodger fan!
Don’t forget his eephus!
What I really want is a GIF where you have his 4 seam overlayed with that change he through to Bethancourt. Because that is how you will see how unfair that change is.
In a vacuum that change doesn’t look all that special, and as was noted in the article it might technically be ‘below average’ intrinsically. But as compared to the movement on his 4 seam, it will probably be devastating.
It will also have the side effect of making his actual 4 seam better, because it will be much harder to ‘sit fastball’ on Kershaw and just hope you make contact.
I’m fairly certain the scouting report on Kersh is ‘he throws a lot of strikes and a lot of early in the count fastballs, so your best chance is to swing early and put the ball in play’.
With this change people will be more off balance and what once was ‘only’ a top 15 4 seam fast ball probably jumps up to a top 10 fastball.
The other interesting thing was that two of those pitches really do look like 2 seam. Could be another way he is going to try to keep people off balance, while still being able to throw a lot of first pitch strike fastballs?
Every year we say ‘oh, but what if Kershaw also had a decent change’. So I don’t really expect it to happen this year. However one of those 2 changes thrown look very good, and the other one looked serviceable [with respects movement, that location didn’t look very good]. As already stated, that change would just be devastating to hitters who are already over-matched.
How much weight should we give the fact that Bethancourt routinely flails helplessly at pitches from big-league pitchers into our evaluation of the impressiveness of Kershaw’s change?
“If Kershaw adds a changeup that looks like the one against Bethancourt, batters are screwed.”
This sentence contains a lot of superfluous information.
“If Kershaw…, batters are screwed.”
There we go.
Kershaw is amazing. It’s been so long since I saw Pedro, Maddux, or Randy Johnson pitch that they’ve become mythical figures to me, but his dominance reminds me of them, even if he hasn’t done it for as long.
He hasn’t done it as long, but excepting his rookie season as a 20yr old, Kershaw has put up numbers over his career (2009-16) that are equal to any eight-year stretch by Pedro, Maddux, or Johnson, let alone during the same 21-28yr seasons.
As a side note, Johnson’s improvement as he aged into his 30s is nuts; if he had been a position player, we would all assume he had been using steroids.
It’s hard to compare Kershaw to Pedro, Maddux, and Johnson because Kershaw has pitched less in basically every season than those three. If Kershaw had pitched more innings, he’d have piled up more WAR but his sterling FIP numbers might be a little weaker because he’d be pitching later into games. By the numbers, Kershaw has 3 of the top 5 FIP seasons in the last 25 years or so, but his WAR numbers were only 6.5, 7.6, and 8.6. On the other hand, Randy Johnson has FIVE seasons with WAR’s higher than anything Kershaw has put together. I don’t think there’s an easy way to compare them, but subjectively it just feels like Kershaw is as dominant as those three and there’s some evidence to back that up.
Side note: Pedro’s 1999 season. O.M.G.
Actually makes no sense.
Pedro from 1995 through 2000 [6 seasons] pitched an average of 219 innings a year.
Kershaw from 2010 through 2016 [6 seasons] pitched an average of 221 innings a year.
Maddux from 1993 through 1998 [6 seasons] pitched an average of 234 innings a year.
Johnson from 1997 through 2002 [6 seasons] pitched an average of 248 innings a year.
So, more average innings pitched than Pedro, a bit less than Maddux, noticeably less than Johnson.
Actually, Kershaw’s six year run got him about 212 IPs (2010-2016 is seven years; his best is 2011-2016). But Pedro’s not a good example anyway, you’re right; he didn’t pile up many innings either and had several looney years too. But Kershaw pitched 132 innings less than Maddux and 216 less than Johnson, so he was definitely less “valuable” even if he wasn’t any less dominant.
The point is just that you could argue that Kershaw has been *more* dominant than Maddux or Johnson based on rate stats. There have been 5 sub-2.00 FIPs in the last 35 years, and Kershaw has THREE of them. But that only got him 22.7 WAR, and he’s behind both Maddux and Johnson on the six-year WAR, although only about 4 WAR behind Maddux and 10 behind Johnson. Should we reward Maddux and Johnson for enduring through more innings, or just look at rate stats? This isn’t a new argument or anything, just thought it was relevant here.
I think it was Johnson’s freakish physical attributes that keep the steroid suspicion away. As a younger player there was always talk that he would be ridiculous if he learned to harness and control his talent.
Of course. And I’m not trying to imply that he was using PEDs. Only that at the time, we were more focused on position players, and specifically hitters who experienced significant power increases, when evidence tells us that use was as prevent amongst pitchers and was used for stamina and recovery as often as for pure strength.
The larger point being that Johnson’s late career was freakishly good.
Yeah, the dude looked like an NBA player. I guess it’s possible he was a PED user, because so many dominant players of that era were, but his height and arms made him a one-of-a-kind pitcher. And he actually had a somewhat normal decline at the very end. It’s not like Clemens who looked like he was declining from 1993-1996, and then suddenly destroyed the league for two years with the Blue Jays.
I was thinking about Kershaw’s changeup around the middle of Spring Training and tried to find a comp for him if he added one by handedness and repertoire. Is it even possible for him to get better, or would he just be a different kind of devastating?
As the article explains, this would probably eliminate the small platoon split that RH batters currently enjoy.
I saw that, but I guess I’m just wondering what that tiny advantage looks like, like what kind of dent it puts in his numbers in the grand scheme of things. Part of it is also trying to comprehend how good a single guy can be.
I know all this information is technically up there, but I think it’s worth specifically noting that both pitches most likely to be actual changeups were thrown ahead in the count with two outs, and no one on. I can’t imagine a better time to experiment. It’s clearly still a work in progress, but I imagine he’ll be able to find other, similarly low leverage spots this year to try working it in.
I’d honestly probably still visit this site just as much if it was only Kershaw, Votto, and Trout articles.